App retargeting continues to be a strategic marketing channel for mobile applications. In recent years, in all mobile verticals (shopping is the leader), the focus on remarketing has increased dramatically. With IDFA traffic continuing to be above 50% of iOS inventory, the app marketing advertisers keep looking to increase their revenue.

What is incrementality measurement in mobile marketing?

Incrementality, as one of the measurements of ad effectiveness, is a way to measure the supplemental business resulting from a marketing tactic or set of marketing tactics. It wouldn't have occurred without a specific interaction, such as an ad view, and that resulted in the desired outcome, such as a conversion. It can be shown by the increase in customers, sales, revenue, or other related metrics.

In performance marketing, key performance indicators (KPIs) are constantly evolving to align with mobile marketing business objects. In order to more efficiently control the quality and return of paid traffic, we have seen KPIs have changed from cost-per-thousand impressions (CPM) to cost-per-click (CPC), and with the introduction of mobile applications, the cost per install ( CPI), cost per operation (CPA), and return on advertising spend (return on advertising spend). The performance KPIs come with new attribution models arrived with them. 

Why is incrementality important in mobile measurement? 

App promotion cost calculation and the performance measure are always a hot topic to be considered by marketers. Wrong metrics may lead to wrong outcomes and incentives, which ultimately lead advertisers to confused about this, "Half of the money I spent on advertising was wasted; the trouble is, I don't know which half it is".

Incrementality testing offers a new way to simplify measurement by comparing the differences in user behavior between two groups that are either exposed or unexposed to ads. When implemented correctly, incrementality offers a definitive way to measure the effectiveness of mobile advertising.

Two goals for app remarketing: 

● To understand the true ROI from marketing campaigns, enabling marketers to correctly attribute the source of performance growth.
● To optimize cross-channel performance and make informed decisions on the next marketing dollar spent.

Incremental analysis is particularly important when it comes to measuring remarketing campaigns, since these users are already engaged with your app, and therefore are more likely to engage again, organically. How can marketers then know whether their paid remarketing efforts are worth the price of these same users are likely to convert again on their own? 

The only way to justify pouring additional funds into re-engagement efforts is by testing their lift.

How do I go about testing and measuring incrementality?

The most accurate way to measure incrementality of media is through testing and experimentation. To measure incrementality, audiences are randomly segmented into test and control cohorts. The difference in conversion rates between the two cohorts effectively gives us incrementality and an accurate read on the marginal incremental contribution of that media channel.

Incrementality in marketing examples:

Retargeting incrementality calculation example
● You withhold a small but statistical group of your audience and do not serve retargeting ads. On average, 10% end up repurchasing your products.
● Test group receives ads, and they repurchase 13% of the time, so the incremental lift is 3% resulting in a 23% incrementality.
● (%CR Test – %CR Control) / %CR Test

Soft surroundings retargeting case study
● Measured’s retargeting experiment exposed that the incremental cost per acquisition CPA(i) was well above CPA targets and what was reported by the vendors. Their largest retargeting vendor by spend was heavily over indexed and serving ads beyond a recommended frequency cap.
● Results: Retargeting budgets were reduced by 52% in the next few months based on findings. The extra budget was shifted to higher-performing prospecting tactics like Facebook. Topline revenue improved 17% MoM while yearly sales comps increased 12%.
Other methods of measurement, such as media mix modeling (MMM) and multi-touch attribution (MTA), cannot measure views or impressions, so you’re essentially just measuring clicks. Incrementality measurement accounts for the impressions and clicks within each of the platforms under test and therefore gives marketers a more accurate view of the true contribution of their media across their entire portfolio.

Leveraging incrementality measurement to value app retargeting

Since mobile app retargeting relies on user ID, the incremental measurement of retargeting marketing campaign is similar to traditional A/B testing. The audience was randomly divided into a treatment group and a control group.

Regarding how to evaluate the incremental improvement of advertising, the Ghost Bids method is one of methods to measure improvement. It is the most accurate and cost-effective method designed for long-term evaluation of advertising effectiveness.

The Ghost Bids is based on Google's "Ghost Ads", which provides a balanced "one-by-one" comparison of the behavior of the treatment group and the control group. However, the Ghost Bids concept is different in that it is adjusted for app retargeting.

Detailed information about Ghost Bids method you can learn: Ghost Bids

KPIs used in incrementality measurement

When measuring the effectiveness of each activity, choosing the right KPI is critical. The right choice depends on the industry and business goals of the app.

The most commonly used metrics:
● Incremental revenue
● Incremental ROAS (iROAS)
● Incremental Conversion
● Incremental Cost Per Operation (iCPA)

Incremental revenue

Incremental Revenue reflects how much of the revenue generated by all targeted users within a campaign is incremental, compared to the behavior of the control group. This KPI captures how much additional revenue a campaign is driving, in addition to organic behavior and other marketing activities.

Incremental ROAS (iROAS)

Incremental ROAS (iROAS) is incremental revenue divided by campaign cost (ad spend). This KPI represents the profitability of the event or its return on investment (ROI).

Incremental Conversion

Similar to incremental revenue, the incremental conversion KPI reflects how many conversions of the users targeted by the campaign are incremental compared to the control group.

Incremental Cost Per Operation (iCPA)

For apps whose business model is not reflected by tracking revenue, or if actual revenue cannot be obtained through data streams, it is most meaningful to check iCPA for goal setting and performance evaluation.



Conversion rate (CVR) is another example where the result depends on the goal set. If the goal is to understand the true value of an advertisement driven by an activity, statistical significance and positive CVR are good performance indicators, even if it may not fully meet KPIs attributed to ROAS.

Things to consider when applying incremental measurement 

● The starts time of your incremental testing

During the off-season, the execution cycle of the campaign will be relatively lower than the usual cycle, and the amount of data generated is small, so the cost of creating a large number of ads is high. The same applies to peak seasons, where performance is usually higher than normal. Therefore, the starts time of incremental testing need to be considered by app promotion marketers.

● Incremental testing and discount campaigns

To get the most accurate incremental test results, avoid using promotional banners or discount codes. Promotions entice users to convert, so the results may capture false reality.

Conclusion

As remarketing budgets grow substantially, understanding the impact of remarketing plans is necessary to justify the scale.

In addition, personalization is a key goal of today’s digital marketers; marketing to a specific audience, controlling the number of times users see ads in that audience, what the information will be, and how these users are constantly moving along the conversion channel and where The concept of interacting with different touch points in the process.

Using person-based attribution methods, the ideal situation is to keep messages personalized, coherent, and consistent. If we look to the future, measuring performance in isolation will not work. Naturally, this leads to the importance of understanding the incremental impact of the entire re-engagement program.